Contents lists available at ScienceDirect

Applied Catalysis B: Environmental

journal homepage: www.elsevier.com/locate/apcatb

π–π Interaction between self-assembled perylene diimide and 3D graphenefor excellent visible-light photocatalytic activity

Jun Yanga, Hong Miaoa, Yunxia Weib, Wenlu Lia, Yongfa Zhua,⁎

a Department of Chemistry, Tsinghua University, Beijing, 100084, PR Chinab College of Chemistry and Chemical Engineering, Lanzhou City University, Lanzhou, 730070, PR China

A R T I C L E I N F O

Keywords:Photocatalystπ–π interactionThree-dimensional structureElectron transfer

A B S T R A C T

A three-dimensional (3D) graphene/self-assembled perylene imide (rGO/PDI) aerogel composite has been suc-cessfully constructed via a simple low-temperature hydrothermal method in this work. The 3D rGO/PDI com-posite exhibited superior visible-light photocatalytic performance. The removal rate was respectively 2.46 timesand 3.33 times higher than that of PDI in static and dynamic catalytic systems. The satisfactory activity is mainlyattributed to the more effective separation of electron–hole pairs. To be specific, the self-assembled PDI nano-fibers and graphene are effectively combined via non-covalent π-π interaction. The π-π interaction enhances thelong-range π-electrons delocalization and electron coupling effects, which is beneficial for the carrier mobilityand the separation of electron–hole pairs. Meanwhile, the three-dimensional structure of graphene provides afast multidimensional channel for electron transfer and effectively improves the adsorption capacity of PDI dueto the large specific surface area. What’s more, the rGO/PDI composite exhibits great stability than the pure PDI.Overall, this work provides a new insight into improving the photocatalytic activity of self-assembled organicmaterials.

1. Introduction

In recent years, the problem of water contamination is becoming anoverwhelming problem all over the world. Photocatalytic oxidationtechnology has shown tremendous potential as a promising avenue inwater treatment because of its environmental friendliness, good cost-eff;ectiveness and high efficiency [1–5]. Among various photocatalysts,organic photocatalytic materials have received widespread attention fortheir flexible designs, various chemical structures, low costs and richorigin of elements [6,7]. At present, the research for organic photo-catalyst is mainly focused on organometallic complexes [8–10] andcovalent organic polymers [11–14]. But their widespread applicationsin the field of photocatalysis are largely limited because of the potentialtoxicity, complicated synthetic process and high cost. Based on this, it isthe key to find ecofriendly organic photocatalyst materials with highphotocatalytic activity.

Recently, self-assembled supramolecular organic materials haveshown great potential in the field of photocatalysis, such as diketo-pyrrolopyrrole (DPP) [15,16], porphyrin (TPP) [17–19], perylenemonoimide (PMI) [20,21] and perylene imide (PDI) [22]. They havebeen applied in light-driven water splitting, photocatalytic degradationand organic photosynthesis. In particular, perylene imide and its

derivatives, as a typical n-type organic semiconductor, have been ap-plied in photocatalysis because of their high photo-thermal stability,excellent charge mobility and electron affinity [23–28]. However, itsuffers from the problems of easy recombination of photo-generatedcarriers, low adsorption capacity and low recycling rate for their poorstability in the photocatalytic process. These defects greatly limit theirphotocatalytic applications. Currently, the modification research ofself-assembled PDI is mainly focused on introducing functional groups[24], forming short-range π-π stacking structure [22], and adjustingelectron acceptor and donor units [26]. These methods are intended toimprove the electron transfer rate via increasing the built-in electricfield of self-assembled PDI. However, the preparation process of thesemethods are complicated and it is negligible for the improvement of itsstability and adsorption capacity. Therefore, it is very important to seekan efficient modification method for organic supramolecular self-as-sembled photocatalysts.

Currently, the three-dimensional (3D) structure carbon materialsexhibit many advantages over their one-dimensional and two-dimen-sional structures because of the larger specific surface area, lower sheetagglomeration and higher thermal/electrical stability [29]. The three-dimensional structure materials have been applied in many fields, suchas battery materials [30], biosensors [31] and photocatalysis [32–34].

https://doi.org/10.1016/j.apcatb.2018.09.003Received 2 July 2018; Received in revised form 28 August 2018; Accepted 2 September 2018

⁎ Corresponding author.E-mail address: zhuyf@mail.tsinghua.edu.cn (Y. Zhu).

Applied Catalysis B: Environmental 240 (2019) 225–233

Available online 05 September 20180926-3373/ © 2018 Elsevier B.V. All rights reserved.

T

In the last few years, a variety of three-dimensional structure photo-catalysts have been successfully synthesized, such as PANI/g-C3N4 [35],g-C3N4/Agar [36], g-C3N4 /SiO2 [37], PANI/TiO2 [38] and TiO2-gra-phene [39]. Among numerous three-dimensional network structurematerials, graphene-based composite photocatalysts have received ex-tensive attention for their high adsorption ability and easy circulation.Importantly, the excellent conductivity of graphene can effectivelypromote the electron transfer and its three-dimensional mesoporousstructure provides a multidimensional channel for mass transfer andelectron transport in catalytic reactions [40,41]. The advantages arevery favorable for photocatalytic reactions. At present, the research on3D graphene-based composites mainly focuses on graphene/inorganicsemiconductors [42,43], but such materials have the disadvantages ofhigh cost, poor interface contact effect, and complicated preparationprocess. Fortunately, constructing the graphene-supramolecular self-assembled organic photocatalysts may be able to overcome these dis-advantages. In addition, considering the deficiencies of PDI as statedabove, it may be a good choice to construct a graphene-supported three-dimensional structure to improve the photocatalytic performance ofPDI.

Based on the above analysis, a three-dimensional graphene/self-assembled perylene imide (rGO/PDI) aerogel composite was success-fully constructed via a simple low-temperature hydrothermal method(Scheme 1). The 3D rGO/PDI composite exhibited a satisfactory pho-tocatalytic activity under visible light irradiation. The excellent activityis mainly attributed to the close π-π interaction between PDI and gra-phene, which greatly facilitated the separation of electron–hole pairs.In addition, the three-dimensional structure effectively improves theadsorption capacity of PDI. In a word, our work provides a new idea forthe modification of self-assembled supramolecular organic photo-catalysts.

2. Experimental section

2.1. The synthesis of rGO/ PDI aerogel composite

The self-assembled perylene diimide (PDI) was synthesized by theprevious work and was described in detail in the supporting file[24,25]. GO was prepared from graphite powder according to Hum-mer’s method [44]. Firstly, a stock solution of 200.0mL of N, N-di(propanoic acid)-perylene-3,4,9,10-tetracarboxylic diimide (5.0 mM)was prepared with addition of 834 μL of triethylamine solution undervigorous stirring. Subsequently, the solution of 4.0M HCl (5.62ml) wasadded into the red stock solution (37.45 ml, containing PDI 100mg)and then stirred for three hours. A certain volume of GO stock solution(1mg/ml) was added under stirring and then ultrasound for 2 h. Then,

a given mass of ascorbic acid was added under stirring followed ul-trasound for another 2 h. The resulting mixture further heated for 4 h at95 °C in a water bath. The as-prepared hydrogels were washed withdistilled water until the pH of filtrate became neutral and then freeze-dried to obtain aerogels, which were denoted as rGO/PDI-X. (rGO/PDI-X, X represented the mass fraction of PDI in the composite system.X=40%,50%, 55%, 57%, 60%, 63%, 65%, 67%, 70%, 80%, 90%)

2.2. Characterization of materials

The X-ray powder diffraction (XRD) data were recorded at roomtemperature by a Bruker D8 Advance X-ray diffractometer at 40 kV and40mA for monochromatized Cu Kα (λ=1.5406 Å) radiation. Themorphologies of the samples were determined with the scanning elec-tron microsopy (FE-SEM, Hitachi SU-8010), the transmission electronmicroscope (TEM, Hitachi HT 7700) and the atomic force microscopy(AFM, SPM-9700). The optical diffuse reflectance spectra were mea-sured by Hitachi U-3010 spectrophotometer and BaSO4 was used as thereference. Fourier transfer infrared (FT-IR) spectra were carried outusing the Bruker V70 spectrometer. Raman spectra were measured onthe microscopic confocal Raman spectrometer (HORIBA HR800) withan excitation of 514 nm laser light. The photoluminescence (PL) spectrawere performed by using an Edinburgh FS5 spectrophotometer with anexcitation wavelength of 612 nm. X-ray photoelectron spectroscopy(XPS) was performed on the ULVAC-PHI-Quantera spectrometer. AMicromeritics ASAP2020 Surface Area and Porosity Analyzer were usedto measure the surface areas of the samples at liquid nitrogen tem-perature (77 K). The photocurrent experiments were performed byusing the electrochemical workstation (CHI-660E, China). The detec-tion of the active species was performed on the JES-FA200 EPRSpectrometer paramagnetic resonance spectrometer.

2.3. Photocatalytic activity test

The photocatalytic performances in static system were evaluated bydegrading phenol (5 ppm) in a multi-tube agitated reactor under visiblelight irradiation (λ > 420 nm). The visible light source was used witha 500W xenon lamp (filter, λ=420 nm) with an average light intensityof 27mW/cm2. In detail, 25mg of photocatalyst was dispersed in aquartz tube containing 50ml phenol aqueous solution (5 ppm) and thesuspension was kept stirring in the dark for 1 h to reach the adsorption-desorption equilibrium on the catalyst surface. Then the suspension wasirradiated by the 500W xenon lamp (λ > 420 nm). After that, 2 mL ofthe suspension was extracted at given intervals. The concentration ofphenol was detected by a high performance liquid chromatography(HPLC) system and the detection wavelength of phenol was 270 nm. In

Scheme 1. Schematic illustration of the synthesis of 3D rGO/PDI composite aerogel.

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

226

addition, the photocatalytic experiments in dynamic system was de-scribed in detail in the supporting file.

3. Results and discussion

3.1. The π−π interaction between rGO and PDI

The preparation process of the 3D rGO/PDI aerogel composite wasillustrated in Scheme 1. Before the hydrothermal reaction, nano PDIand graphene oxide (GO) aqueous solution were mixed sufficiently,which enabled the PDI to be adsorbed eff ;ectively on the surface of GOdue to the π-π interactions between them [45–47]. Subsequently, themost of hydrophilic oxygen-containing functional groups of GO wereremoved via the hydrothermal reaction to form reduced GO (rGO),which could recover the π-π conjugation between layers [48]. Mean-while the residual hydrophilic groups could absorb water via hydrogenbonding, which can prevent parallel stacking of rGO sheets, helping toform the 3D structure [49].

Fig. 1 shows the XRD patterns of the as-prepared samples. It can beseen that rGO shows a broad peak of (002) at 2θ=22.4°, which cor-responds to a layer-to-layer spacing of 3.5 Å. The diffraction peaks ofrGO are wide and weak, which indicates that interlayer peeling occurswith the reduction of graphene oxide and it may form a thin layer oreven a single layer. As for the self-assembled PDI, it has a clear π-πstacking peak at 2θ=25.9°, corresponding to its lattice spacing of 3.4 Å[50]. It is interesting that the typical d-spacing of π-π stacking of PDI issimilar to the layer-to-layer spacing of graphene, which provides auseful condition for the generation of π-π interactions between them.The enlargement of electron delocalization effect is beneficial to facil-itate charge transfer. For all the composites, the intensities of the dif-fraction peaks of PDI increase with its increasing mass fraction. In ad-dition, it can be seen that the diffraction peaks of the composites shiftedtoward low angles compared to the self-assembled PDI, which indicatesthe existence of π-π interactions between them.

To further illustrate the π-π interaction between rGO and PDI,Raman and infrared measurements were performed. As shown in theFig. 2(A), it can be seen that there are two main peaks of graphene,namely the disordered peak (marked as D peak, 1350 cm−1) and thegraphite peak (marked as G peak, 1586 cm−1). It is obvious that the ID/IG value of rGO (1.25) is higher than that of GO (0.94), indicating thatthe oxygen-containing groups are largely reduced [51,52]. As for PDI,the peak of 1582 cm-1 is attribute to the stretching vibration of C]C/CeC, which is sensitive to the π-π stacking interactions. However, the

1296 cm−1 peak is insensitive to the π-π stacking effect correspondingto the CH in plane bending. Therefore, the intensity ratio of1582 cm−1/1296 cm−1 can be used to evaluate the degree of π–πstacking [52]. It turns out that the intensity ratio of rGO/PDI is 0.86,which is even higher than that of PDI (0.69). This indicates that theaddition of graphene does not affect the π-π stacking of PDI and thesimilar stack spacing even promotes the degree of ordered π-π stacking.The coplanar π-π stacking along the long axis of the PDI nanofibers isformed between the graphene and PDI, which enhances the long-rangeπ-electrons delocalization and electron coupling effects, and then it isbeneficial for the carrier mobility [53,54]. In addition, the Raman peakof the composite at 1561 cm−1 shows a shift compared to that of PDI(1582 cm−1), which further indicates that the existence of π-π inter-action between graphene and PDI.

Fig. 2(B) shows the Fourier transform infrared (FT-IR) spectra of PDIand rGO/PDI composite. The peak of PDI at 1651 cm −1 is attributed tothe C]C stretching vibrations [24]. Interestingly, it can be clearly seenthat the infrared absorption peaks of rGO/PDI has a red shift comparedto pure PDI, which indicates the increase of the conjugated structure inrGO/PDI. It is further demonstrated that the existence of π-π interactionbetween graphene and PDI.

To further illustrate the chemical state of the sample, XPS wascarried out. As shown in Fig. 3, the C 1 s of rGO exhibits binding energypeaks at 284.7 eV, 286.7 eV, and 289.1 eV, which correspond to sp2

carbon C]C, eCeO, and C]O, respectively [55–58]. It indicates thatthere are still certain amounts of the oxygen-containing functionalgroups on the surface of rGO, and their hydrophilicity is favorable forthe photocatalytic reaction. The extra CeN peak in rGO/PDI is mainlyderived from PDI. It can be seen that the C]C peak at 284.4 eV of rGO/PDI is shifted by 0.2 eV toward low binding energy compared to purePDI (284.6 eV), which is mainly attributed to the increase of electrondensity induced by π-π interaction between graphene and PDI. It in-dicates that the addition of rGO extends the π-conjugation of PDIthrough the π−π interaction and obviously increases the electrondensity [55]. The peak fitting analysis data of C 1 s are shown in tableS1. It can be seen that the C]C content of rGO/PDI (∼73.87%) ishigher than the individual PDI (∼61.62%). The result clearly indicatesthat the introduction of rGO increases the C]C bond intensity. In ad-dition, the Fig. 3(D) shows that the N 1 s peak of rGO/PDI (399.2 eV) isshifted by 0.4 eV toward low binding energy compared to pure PDI(399.6 eV), which further implies the existence of π-π interaction be-tween graphene and PDI.

3.1.1. The three-dimensional network structureFig. 4 shows the SEM images of samples and the TEM is shown in

the supporting documents (Fig. S1). It can be seen that the pure PDIgenerates obvious agglomeration (Fig. 4A) and it exhibits a typical one-dimensional nanofiber structure (Fig. S1A). The rGO presents a three-dimensional network structure and the ultrathin graphene sheets cross-link together (Figs. 4B and S1 B). The SEM pictures of rGO/PDI com-posite (Fig. 4C and D) show that the self-assembled PDI nanofibers aredispersed in the three-dimensional network structure of the graphene.The results indicate that graphene does not affect the orderly self-as-sembly of PDI and it in turn improves the dispersion of PDI. More im-portantly, the 3D structure can provide multi-dimensional channels forelectron transport, which is very favorable for photocatalytic reactions[39]. The high-resolution transmission of rGO/PDI (Fig. S1 D) furthersuggests that PDI is sufficiently integrated with graphene sheet. Theclose interfacial effect is mainly attributed to the non-covalent π-π in-teraction between graphene and PDI, which is beneficial to the electrontransfer.

The atomic force is further investigated as shown in Fig. 5. Theresult shows that graphene presents a very thin layer with a thickness of0.3∼0.4 nm (Fig. 5A). The self-assembled PDI shows the one-dimen-sional nanofiber morphology (Fig. 5B), which is stacked in severallayers and the thickness of a single sheet is mainly distributed around

Fig. 1. XRD patterns of rGO, PDI and rGO/PDI-X (rGO/PDI-X, X represents themass fraction of PDI in the composite system).

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

227

7 nm–10 nm. It indicates that the individual PDI is prone to agglom-eration, which is accordant with the result of SEM in Fig. 4A. As forrGO/PDI, the three-dimensional photograph (Fig. 5C) vividly showsthat rGO and PDI are closely knit together. The layer thickness of rGO/PDI is about 7 nm, which is similar to the monolayer of PDI. This furtherindicates that graphene improves the dispersion of PDI.

For rGO/PDI composite, the separation of photo-generated carriersmainly occurs at the interface between graphene and the central cata-lyst PDI, so increasing the interfacial contact will facilitate the se-paration of photo-generated carriers. As shown in SEM, TEM and AFM,it can be concluded that the three-dimensional network structure im-proves the dispersion of PDI and facilitates the close interface contacteffect, which is essential for the improvement of photocatalytic per-formance.

In addition, Fig. S2 shows the UV–vis diffuse reflectance spectra ofthe samples. The absorption edges of PDI is about 738 nm, corre-sponding to the bandgap of 1.68 eV, which indicates that it can be

easily excited by visible light. It is obvious that the addition of grapheneenhances the ability of visible light absorption of PDI and induces sig-nificant red shift, which is also favorable for the photocatalytic reac-tion. Fig. S3 presents the N2 adsorption-desorption isotherms of thesamples, the result shows that the specific surface area of rGO/PDIcomposites increases at different degree compared to pure PDI. Thelarger specific surface area can promote the adsorption ability of thecatalysts and provide more reactive sites for the photocatalytic reac-tion. Thus it can be seen that the advantages of the three-dimensionalstructure in the photocatalytic degradation are manifold.

3.2. Photocatalytic activity in static system

The adsorption capabilities of samples are shown in Fig. S4(Supporting information). As shown in Fig. S4, it took about one hour toget adsorption equilibrium and the adsorption capacity of rGO/PDIcomposite was much better than that of pure PDI. This indicates that

Fig. 2. (A) Raman spectra and (B) FT-IR spectra of PDI and rGO/PDI aerogel composite.

Fig. 3. The high-resolution regional spectra of C 1 s orbital of (A) rGO, (B) PDI and (C) rGO/PDI and (D) the XPS spectra of N1 s orbital of PDI and rGO/PDI.

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

228

the three-dimensional network structure is beneficial to increase theadsorption and enrichment ability of the pure PDI, which is critical tothe subsequent photocatalytic reaction.

The visible-light photocatalytic performance of samples is evaluatedafter adsorption equilibrium. As shown in Fig. 6(A), the pure PDIshowed a relatively low activity in degrading phenol. Interestingly, the

photocatalytic activity of rGO/PDI composite was enhanced graduallywith the increasing mass ratio of PDI. When the mass fraction of PDIreached sixty percent, it exhibited the highest photocatalytic activity.The apparent rate constant k of rGO/PDI-60% is 0.465 h−1, which isalmost 2.46 times as high as that of PDI (0.189 h−1). As shown inFig. 6(B), the peak located at 3.169min corresponds to phenol and the

Fig. 4. SEM images of (A) PDI, (B) rGO and (C, D) rGO/PDI-60% composite aerogel.

Fig. 5. Atomic force micrograph (AFM) of (A) rGO, (B) PDI, and (C) rGO/PDI-60%.

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

229

HPLC results showed that the concentration of phenol gradually de-creased with continuous irradiation. Few intermediate species and PDIwere detected, indicating that the dissolution of PDI in the compositestructure was very few and phenol was mineralized to CO2.

The stability of PDI is an important issue in the photodegradationreaction. As shown in Fig. 6(C), the rGO/PDI composite remainedhigher stability than the pure PDI in neutral environment. Furthermore,the catalytic stability of rGO/PDI was also investigated in the acid andalkaline environment (Fig. 6D). The results indicated that the rGO/PDIcomposite has higher stability in the acid environment than in theneutral and alkaline environment. The decreased activity at neutral andalkaline solution could be attributed to the depolymerization of PDIduring the photodegradation [24].

The activity of rGO/PDI composite was further investigated com-pared with other reported typical visible-light photocatalysts. As shownin Fig. S6, for phenol degradation, the activity of rGO/PDI-60% is 5.93,16.06, 3.15, 2.11 times higher than that of Bi2WO6, g-C3N4, bulk PDI,and commercial self-assembly PDI, respectively. In addition, 2, 4-Dichlorophenol have also been used as the probe molecule to evaluatethe activity. The results showed that rGO/PDI also had the best activitythan other catalysts.

3.3. The synergetic effect of adsorption and photocatalysis in DynamicSystem

The synergetic effect of adsorption and photocatalysis was in-vestigated in a dynamic system, namely, the peristaltic pump and thelight source were turned on at the same time in the photocatalytic test[38]. As shown in Fig. 7(A), the degradation rate of rGO/PDI compositefor methylene blue (MB) degradation was 3.33 times as high as that ofPDI after light irradiation for ten hours. This fully demonstrates thatrGO/PDI composite has the synergistic effect between adsorption andphotocatalysis in the photocatalytic degradation. As for the rGO/PDIcomposite, the three-dimensional network structure of graphene im-proves the adsorption ability for pollutants. It is worth noting that thephotocatalytic performance of PDI is significantly reduced with the

continuous reaction, while the rGO/PDI composite exhibits satisfactoryphotocatalytic stability. It further indicates that the addition of gra-phene improves the stability of PDI. The maximum absorption peak ofMB can be seen at 663 nm, which gradually decreases from 0min to80min, which indicates that the concentration of MB decreases withirradiation time.

3.4. Mechanism of photocatalytic activity enhancement

The transient photocurrent and photoluminescence characterizationwere used to analyze the separation and migration efficiency of photo-generated carriers. As shown in Fig. 8(A), PDI and rGO/PDI compositeboth showed a relatively stable instantaneous photocurrent signal.However, the photocurrent intensity of rGO/PDI (3.2 μA) is about 13times higher than that of PDI (0.25 μA). It indicated that the addition ofgraphene facilitated the separation and transmission of photo-gener-ated carriers. It is not only attributed to the excellent conductivity ofgraphene, but more importantly, the close π-π interaction betweengraphene and PDI promotes the electron delocalization and the three-dimensional structures provides a fast channel for the electron transfer.As shown in Fig. 8(B), the self-assembled PDI has a strong emissionpeak at 733 nm, which is mainly due to the recombination of photo-generated electrons and holes. However, the fluorescence intensity ofthe rGO/PDI composite was obviously weakened, which further de-monstrated that the separation and transmission of photon-generatedcarriers are enhanced after the addition of graphene.

In addition, the rGO/PDI-60% composite shows higher photo-catalytic performance than the physical mixture (the mass fraction ofPDI is 60%) as shown in Fig. S7, which indicates that the π-π interactionbetween graphene and PDI is the main factor for the photocatalyticactivity enhancement.

Active species plays a crucial role in the photocatalytic reaction, sowhat kind of active species dominates the photo-degradation reaction?The free radical trapping experiment was conducted. The experimentaldetails were shown in the supporting documents. As shown in Fig. 9,formic acid, tert-butyl alcohol, and benzoquinone are used as the

Fig. 6. (A) The degradation rate constants k ofsamples for phenol degradation; (B) HPLCspectrums of phenol solution degraded byrGO/PDI-60% composite; (C) Recycling tests ofPDI and rGO/PDI-60% for phenol degradationin the neutral environment; (D) Recycling testsof rGO/PDI-60% for phenol degradation in theacid (pH=4.8) and alkaline (pH=8.7) en-vironment.

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

230

trapping agents for holes (h+), hydroxyl radicals (OH), and superoxideradicals (O2

%−), respectively. The result shows that the photo-de-gradation of phenol was apparently inhibited when formic acid wasadded. This phenomenon gives evidence that the degradation of phenolis dominated by the direct hole oxidation. On the other hand, it can beseen that the photocatalytic performance reduced to some extent afteradding tert-butyl alcohol and benzoquinone, indicating OH and O2

%-also participate in the photocatalytic degradation. The electron spinresonance (ESR) measurement were further performed to detect OHand O2

%-. As shown in Fig. S8, there was no signal response in the ab-sence of light. However, obvious signals appear after illumination,which further indicated that OH and O2

%- indeed play a role in thephotocatalytic process.

Based on the above discussion, the photocatalytic degradation me-chanism of the aerogel composite was proposed, which was shown inFig. 10. Firstly, the self-assembled PDI and 3D graphene are effectivelycombined via non-covalent π-π interaction. The central catalytic com-ponent PDI will be excited under the irradiation of visible light as itsnarrow bandgap (1.68 eV). And then the electrons and holes can berespectively generated in the conduction band (−0.17 eV vs. NHE) andvalence band (1.52 eV vs. NHE) [24]. The enlargement of electron de-localization effect induced by π-π interaction is beneficial to facilitatecharge transfer. The excellent electrical conductivity of graphene andthe three-dimensional structure also provide a fast multidimensionalchannel for electron transfer. So the photo-generated electrons on theconduction band of PDI can easily transferred to the graphene sheet,which will further react with dissolved oxygen in water to form su-peroxide radicals. Superoxide radicals have strong oxidation capacity,which can decompose or even mineralize organic pollutants. The high-speed electron transfer promotes the separation efficiency of carriers,making the photo-generated holes effectively participate in the photo-catalytic reaction. On the one hand, the photo-generated holes can di-rectly mineralize organic contaminants. On the other hand, h+ can

combine with hydroxyl in water to generate OH according to the ESRresult and then organic pollutants can be decomposed by OH. Thephotocatalytic mechanism further indicates that the π-π interaction,three-dimensional structure and excellent conductivity of grapheneplay a crucial synergetic role in the electron transfer process of pho-tocatalytic reactions.

4. Conclusion

In conclusion, a three-dimensional graphene/self-assembled per-ylene imide (rGO/PDI) aerogel composite has been successfully con-structed via a simple low-temperature hydrothermal method. Grapheneand PDI are effectively combined via non-covalent π-π interaction. The

Fig. 7. (A) Adsorption and photocatalytic de-gradation synergistic removal of MB by PDIand rGO/PDI-60% composite with 3.5 mL/minflow rate in dynamic system; (B) the UV ab-sorbance of methylene blue with time (from0min to 80min). (For interpretation of thereferences to colour in this figure legend, thereader is referred to the web version of thisarticle).

Fig. 8. (A) Photocurrents of PDI and rGO/PDI-60% under the irradiation of visible light and (B) Photoluminescence spectra of pure PDI and rGO/PDI composites.

Fig. 9. Effects of different scavengers on phenol degradation in the presence ofrGO/PDI-60% under visible light irradiation.

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

231

three-dimensional aerogel composite exhibits satisfactory photo-catalytic performance and stability in static and dynamic catalyticsystems. On the one hand, the close π-π interaction and the excellentconductivity of graphene facilitate the separation of charge carriers. Onthe other hand, the three-dimensional network structure improves thedispersion of PDI and provides a high-speed multidimensional channelfor electron transfer. Finally, hole (h+) mainly dominates the photo-degradation reaction.

Acknowledgements

This work was partly supported by Chinese National ScienceFoundation (21437003, 21673126, 21761142017, 21621003) andCollaborative Innovation Center for Regional Environmental Quality.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in theonline version, at doi:https://doi.org/10.1016/j.apcatb.2018.09.003.

References

[1] C. Chen, W. Ma, J. Zhao, Semiconductor-mediated photodegradation of pollutantsunder visible-light irradiation, Chem. Soc. Rev. 39 (2010) 4206–4219.

[2] L. Zhao, X. Chen, X. Wang, Y. Zhang, W. Wei, Y. Sun, M. Antonietti, M.M. Titirici,One-step solvothermal synthesis of a carbon@TiO(2) dyade structure effectivelypromoting visible-light photocatalysis, Adv. Mater. 22 (2010) 3317–3321.

[3] Y. Wang, R. Shi, J. Lin, Y. Zhu, Enhancement of photocurrent and photocatalyticactivity of ZnO hybridized with graphite-like C3N4, Energy Environ. Sci. 4 (2011)2922.

[4] H. Liu, J.B. Joo, M. Dahl, L. Fu, Z. Zeng, Y. Yin, Crystallinity control of TiO2 hollowshells through resin-protected calcination for enhanced photocatalytic activity,Energy Environ. Sci. 8 (2015) 286–296.

[5] J. Yang, R. Hu, W. Meng, Y. Du, A novel p-LaFeO3/n-Ag3PO4 heterojunction pho-tocatalyst for phenol degradation under visible light irradiation, Chem. Commun.(Camb.) 52 (2016) 2620–2623.

[6] Y.B. Huang, J. Liang, X.S. Wang, R. Cao, Multifunctional metal-organic frameworkcatalysts: synergistic catalysis and tandem reactions, Chem. Soc. Rev. 46 (2017)126–157.

[7] A. Kudo, Y. Miseki, Heterogeneous photocatalyst materials for water splitting,Chem. Soc. Rev. 38 (2009) 253–278.

[8] C. Wang, Z. Xie, K.E. deKrafft, W. Lin, Doping metal-organic frameworks for wateroxidation, carbon dioxide reduction, and organic photocatalysis, J. Am. Chem. Soc.133 (2011) 13445–13454.

[9] J. Qiu, X. Zhang, Y. Feng, X. Zhang, H. Wang, J. Yao, Modified metal-organic fra-meworks as photocatalysts, Appl. Catal. B 231 (2018) 317–342.

[10] C.-C. Wang, J.-R. Li, X.-L. Lv, Y.-Q. Zhang, G. Guo, Photocatalytic organic pollutants

degradation in metal–organic frameworks, Energy Environ. Sci. 7 (2014)2831–2867.

[11] J. Zhang, X. Chen, K. Takanabe, K. Maeda, K. Domen, J.D. Epping, X. Fu,M. Antonietti, X. Wang, Synthesis of a carbon nitride structure for visible-lightcatalysis by copolymerization, Angew. Chem. Int. Ed. Engl. 49 (2010) 441–444.

[12] K. Schwinghammer, B. Tuffy, M.B. Mesch, E. Wirnhier, C. Martineau, F. Taulelle,W. Schnick, J. Senker, B.V. Lotsch, Triazine-based carbon nitrides for visible-light-driven hydrogen evolution, Angew. Chem. Int. Ed. Engl. 52 (2013) 2435–2439.

[13] L. Kong, Y. Ji, Z. Dang, J. Yan, P. Li, Y. Li, S.F. Liu, G-C3N4 loading black phos-phorus quantum dot for efficient and stable photocatalytic H2 generation undervisible light, Adv. Funct. Mater. (2018) 1800668.

[14] H. Bian, Y. Ji, J. Yan, P. Li, L. Li, Y. Li, S. Frank Liu, In situ synthesis of few-layeredg-C3N4 with vertically aligned MoS2 loading for boosting solar-to-Hydrogen gen-eration, Small 14 (2018).

[15] G.S. Thool, K. Narayanaswamy, A. Venkateswararao, S. Naqvi, V. Gupta, S. Chand,V. Vivekananthan, R.R. Koner, V. Krishnan, S.P. Singh, Highly directional 1D su-pramolecular assembly of new diketopyrrolopyrrole-based gel for organic solar cellapplications, Langmuir 32 (2016) 4346–4351.

[16] C. Fu, P.J. Beldon, D.F. Perepichka, H-bonding control of supramolecular orderingof diketopyrrolopyrroles, Chem. Mater. 29 (2017) 2979–2987.

[17] P. Guo, P. Chen, W. Ma, M. Liu, Morphology-dependent supramolecular photo-catalytic performance of porphyrin nanoassemblies: from molecule to artificialsupramolecular nanoantenna, J. Mater. Chem. 22 (2012) 20243–20249.

[18] C. Zhang, P. Chen, H. Dong, Y. Zhen, M. Liu, W. Hu, Porphyrin supramolecular 1Dstructures via surfactant-assisted self-assembly, Adv. Mater. 27 (2015) 5379–5387.

[19] M. Más-Montoya, R.A. Janssen, The eff ;ect of H-and J-aggregation on the photo-physical and photovoltaic properties of small thiophene–pyridine–dpp moleculesfor bulk-heterojunction solar cells, Adv. Funct. Mater. 27 (2017).

[20] A.S. Weingarten, R.V. Kazantsev, L.C. Palmer, M. McClendon, A.R. Koltonow,A.P. Samuel, D.J. Kiebala, M.R. Wasielewski, S.I. Stupp, Self-assembling hydrogelscaffolds for photocatalytic hydrogen production, Nat. Chem. 6 (2014) 964–970.

[21] A.S. Weingarten, R.V. Kazantsev, L.C. Palmer, D.J. Fairfield, A.R. Koltonow,S.I. Stupp, Supramolecular packing controls H2 photocatalysis in chromophoreamphiphile hydrogels, J. Am. Chem. Soc. 137 (2015) 15241–15246.

[22] D. Liu, J. Wang, X. Bai, R. Zong, Y. Zhu, Self-assembled PDINH supramolecularsystem for photocatalysis under visible light, Adv. Mater. 28 (2016) 7284–7290.

[23] W. Wei, D. Liu, Z. Wei, Y. Zhu, Short-range π–π stacking assembly on P25 TiO2

nanoparticles for enhanced visible-light photocatalysis, ACS Catal. 7 (2016)652–663.

[24] J. Wang, W. Shi, D. Liu, Z. Zhang, Y. Zhu, D. Wang, Supramolecular organic na-nofibers with highly efficient and stable visible light photooxidation performance,Appl. Catal. B 202 (2017) 289–297.

[25] K. Zhang, J. Wang, W. Jiang, W. Yao, H. Yang, Y. Zhu, Self-assembled perylenediimide based supramolecular heterojunction with Bi2WO6 for efficient visible-light-driven photocatalysis, Appl. Catal. B 232 (2018) 175–181.

[26] W. Wei, Z. Wei, D. Liu, Y. Zhu, Enhanced visible-light photocatalysis via back-electron transfer from palladium quantum dots to perylene diimide, Appl. Catal. B230 (2018) 49–57.

[27] J. Wang, D. Liu, Y. Zhu, S. Zhou, S. Guan, Supramolecular packing dominantphotocatalytic oxidation and anticancer performance of PDI, Appl. Catal. B 231(2018) 251–261.

[28] Z. Zhang, J. Wang, D. Liu, W. Luo, M. Zhang, W. Jiang, Y. Zhu, Highly efficientorganic photocatalyst with full visible light Spectrum through pi-pi stacking ofTCNQ-PTCDI, ACS Appl. Mater. Interfaces 8 (2016) 30225–30231.

Fig. 10. Schematic diagram of electron–hole pairs’ separation process and the photocatalytic process under visible light (λ > 420 nm).

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

232

[29] A. Dasgupta, L.P. Rajukumar, C. Rotella, Y. Lei, M. Terrones, Covalent three-di-mensional networks of graphene and carbon nanotubes: synthesis and environ-mental applications, Nano Today 12 (2017) 116–135.

[30] Q. Li, B. Quan, W. Li, J. Lu, J. Zheng, X. Yu, J. Li, H. Li, Electro-plating and strippingbehavior on lithium metal electrode with ordered three-dimensional structure,Nano Energy 45 (2018) 463–470.

[31] X. Wang, A. Liu, Y. Xing, H. Duan, W. Xu, Q. Zhou, H. Wu, C. Chen, B. Chen, Three-dimensional graphene biointerface with extremely high sensitivity to single cancercell monitoring, Biosens. Bioelectron. 105 (2018) 22–28.

[32] X. Wang, Q. Liu, Q. Yang, Z. Zhang, X. Fang, Three-dimension g-C3N4 aggregatecomposed of hollow bubbles with high activity for photocatalytic degradation oftetracycline, Carbon 136 (2018) 103–112.

[33] B. Qiu, M. Xing, J. Zhang, Mesoporous TiO2 nanocrystals grown in situ on grapheneaerogels for high photocatalysis and lithium-ion batteries, J. Am. Chem. Soc. 136(2014) 5852–5855.

[34] C. Hou, Q. Zhang, Y. Li, H. Wang, P25-graphene hydrogels: room-temperaturesynthesis and application for removal of methylene blue from aqueous solution, J.Hazard. Mater. 205-206 (2012) 229–235.

[35] W. Jiang, W. Luo, R. Zong, W. Yao, Z. Li, Y. Zhu, Polyaniline/carbon nitride na-nosheets composite hydrogel: a separation-free and high-efficient photocatalystwith 3D hierarchical structure, Small 12 (2016) 4370–4378.

[36] M. Zhang, W. Jiang, D. Liu, J. Wang, Y. Liu, Y. Zhu, Y. Zhu, Photodegradation ofphenol via C3N4-agar hybrid hydrogel 3D photocatalysts with free separation, Appl.Catal. B 183 (2016) 263–268.

[37] M. Zhang, W. Luo, Z. Wei, W. Jiang, D. Liu, Y. Zhu, Separation free C3N4/SiO2

hybrid hydrogels as high active photocatalysts for TOC removal, Appl. Catal. B 194(2016) 105–110.

[38] W. Jiang, Y. Liu, J. Wang, M. Zhang, W. Luo, Y. Zhu, Separation-free Polyaniline/TiO2 3D hydrogel with high photocatalytic activity, Adv. Mater. Interfaces 3 (2016)1500502.

[39] X. Chen, Q. Chen, W. Jiang, Z. Wei, Y. Zhu, Separation-free TiO2-graphene hydrogelwith 3D network structure for efficient photoelectrocatalytic mineralization, Appl.Catal. B 211 (2017) 106–113.

[40] B. Qiu, M. Xing, J. Zhang, Recent advances in three-dimensional graphene basedmaterials for catalysis applications, Chem. Soc. Rev. 47 (2018) 2165–2216.

[41] F. Chen, W. An, L. Liu, Y. Liang, W. Cui, Highly efficient removal of bisphenol A by athree-dimensional graphene hydrogel-AgBr@rGO exhibiting adsorption/photo-catalysis synergy, Appl. Catal. B 217 (2017) 65–80.

[42] C. Mu, Y. Zhang, W. Cui, Y. Liang, Y. Zhu, Removal of bisphenol A over a separationfree 3D Ag3 PO4 -graphene hydrogel via an adsorption-photocatalysis synergy,Appl. Catal. B 212 (2017) 41–49.

[43] F. Chen, W. An, Y. Li, Y. Liang, W. Cui, Fabricating 3D porous PANI/TiO2–graphenehydrogel for the enhanced UV-light photocatalytic degradation of BPA, Appl. Surf.Sci. 427 (2018) 123–132.

[44] W.S. Hummers Jr, R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc.80 (1958) 1339.

[45] P. He, J. Sun, S. Tian, S. Yang, S. Ding, G. Ding, X. Xie, M. Jiang, Processableaqueous dispersions of graphene stabilized by graphene quantum dots, Chem.Mater. 27 (2014) 218–226.

[46] R. Wang, K.-Q. Lu, F. Zhang, Z.-R. Tang, Y.-J. Xu, 3D carbon quantum dots/gra-phene aerogel as a metal-free catalyst for enhanced photosensitization efficiency,Appl. Catal. B 233 (2018) 11–18.

[47] J. Qian, J. Yan, C. Shen, F. Xi, X. Dong, J. Liu, Graphene quantum dots-assistedexfoliation of graphitic carbon nitride to prepare metal-free zero-dimensional/two-dimensional composite photocatalysts, J. Mater. Sci. 53 (2018) 12103–12114.

[48] Y. Xu, K. Sheng, C. Li, G. Shi, Self-assembled graphene hydrogel via a one-stephydrothermal process, ACS Nano 4 (2010) 4324–4330.

[49] W. Lv, C. Zhang, Z. Li, Q.H. Yang, Self-assembled 3D graphene monolith from so-lution, J. Phys. Chem. Lett. 6 (2015) 658–668.

[50] A. Datar, K. Balakrishnan, L. Zang, One-dimensional self-assembly of a water so-luble perylene diimide molecule by pH triggered hydrogelation, Chem. Commun.(Camb.) 49 (2013) 6894–6896.

[51] X. Yang, H. Cui, Y. Li, J. Qin, R. Zhang, H. Tang, Fabrication of Ag3PO4-Graphenecomposites with highly efficient and stable visible light photocatalytic performance,ACS Catal. 3 (2013) 363–369.

[52] M. Angelella, C. Wang, M.J. Tauber, Resonance Raman spectra of a perylene bis(dicarboximide) chromophore in ground and lowest triplet states, J. Phys. Chem. A117 (2013) 9196–9204.

[53] M.X. Zhang, G.J. Zhao, Modification of n-type organic semiconductor performanceof perylene diimides by substitution in different positions: two-dimensional pi-stacking and hydrogen bonding, ChemSusChem 5 (2012) 879–887.

[54] C. Xavier, C. Jérôme, F. Rainer, K.K. Okudaira, V. Lemaur, A. Crispin, G. Kestemont,M. Lehmann, M. Fahlman, R. Lazzaroni, Y. Geerts, G. Wendin, N. Ueno, J. Bre´das,W.R. Salaneck, Electronic delocalization in discotic liquid crystals: a joint experi-mental and theoretical study, J. Am. Chem. Soc. 126 (2004) 11889.

[55] P. Karthik, R. Vinoth, P. Zhang, W. Choi, E. Balaraman, B. Neppolian, Π–π inter-action between metal–Organic framework and reduced graphene oxide for visible-light photocatalytic H2 production, Acs Appl. Energy Mater. 1 (5) (2018)1913–1923.

[56] D. Xu, B. Cheng, W. Wang, C. Jiang, J. Yu, Ag2CrO4/g-C3N4/graphene oxide ternarynanocomposite Z-scheme photocatalyst with enhanced CO2 reduction activity,Appl. Catal. B 231 (2018) 368–380.

[57] Q. Cheng, J. Xu, T. Wang, L. Fan, R. Ma, X. Yu, J. Zhu, Z. Xu, B. Lu, Double quantumdots decorated 3D graphene flowers for highly efficient photoelectrocatalytic hy-drogen production, Appl. Surf. Sci. 422 (2017) 528–535.

[58] Y. Zhao, X. Zhang, C. Wang, Y. Zhao, H. Zhou, J. Li, H. Jin, The synthesis ofhierarchical nanostructured MoS2/Graphene composites with enhanced visible-light photo-degradation property, Appl. Surf. Sci. 412 (2017) 207–213.

J. Yang et al. Applied Catalysis B: Environmental 240 (2019) 225–233

233